tssims

This package contains a small set of simulation functions. The fractional Brownian motion and FARIMA functions are wrappers for the corresponding functions in the fArma and fracdiff packages, respectively.

The simulation model is created separately from the function. This was done to enable easier randomization of the model parameters using the jitter_params function.

Example

Calling the base simulation function generates the model parameters. The function gen returns a function which takes a single length parameter.

# The model
alpha = 0.8
mod <- weierstrass(alpha)
# The function
f <- gen(mod)
plot(f(200), type = 'l', 
             lwd = 2, 
             col = "burlywood4", 
             main = paste0("Random Weierstrass with alpha=", alpha))

Jitter

The parameter model for the function can be perturbed using the jitter_params function.

mod <- weierstrass(0.8)
# Create 6 versions with perturbed parameters
fs <- replicate(6, gen(jitter_params(mod))(200))
plot.ts(fs, 
        main = "Random Weierstrass with perturbed alpha parameter",
        col = "burlywood4")

Installation

install.packages("devtools") 
install_github("nwaff/tssims")